4-D generative model for PET/MRI reconstruction

Stefano Pedemonte; Alexandre Bousse; Brian F. Hutton; Simon Arridge; Sebastien Ourselin

doi:10.1007/978-3-642-23623-5_73

4-D generative model for PET/MRI reconstruction

Stefano Pedemonte^*, Alexandre Bousse, Brian F. Hutton, Simon Arridge, Sebastien Ourselin

^*Corresponding author for this work

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

15 Citations (Scopus)

Abstract

We introduce a 4-dimensional joint generative probabilistic model for estimation of activity in a PET/MRI imaging system. The model is based on a mixture of Gaussians, relating time dependent activity and MRI image intensity to a hidden static variable, allowing one to estimate jointly activity, the parameters that capture the interdependence of the two images and motion parameters. An iterative algorithm for optimisation of the model is described. Noisy simulation data, modeling 3-D patient head movements, is obtained with realistic PET and MRI simulators and with a brain phantom from the BrainWeb database. Joint estimation of activity and motion parameters within the same framework allows us to use information from the MRI images to improve the activity estimate in terms of noise and recovery.

Original language	English
Title of host publication	Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings
Pages	581-588
Number of pages	8
Edition	Part 1
DOIs	https://doi.org/10.1007/978-3-642-23623-5_73
Publication status	Published - 2011
Event	14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011 - Toronto, ON, Canada Duration: 18 Sept 2011 → 22 Sept 2011

Publication series

Name	Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume	6891
ISSN (Print)	03029743
ISSN (Electronic)	16113349

Other

Other	14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011
Country/Territory	Canada
City	Toronto, ON
Period	18/09/11 → 22/09/11

Keywords

Bayesian Networks
Emission Tomography
Molecular Imaging
Motion correction
Multi-modality

Access to Document

10.1007/978-3-642-23623-5_73

Cite this

Pedemonte, S., Bousse, A., Hutton, B. F., Arridge, S., & Ourselin, S. (2011). 4-D generative model for PET/MRI reconstruction. In Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings (Part 1 ed., pp. 581-588). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6891). https://doi.org/10.1007/978-3-642-23623-5_73

Pedemonte, Stefano ; Bousse, Alexandre ; Hutton, Brian F. et al. / 4-D generative model for PET/MRI reconstruction. Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings. Part 1. ed. 2011. pp. 581-588 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).

@inproceedings{b00d13ad5aac48f196b4a85e4b285d5f,

title = "4-D generative model for PET/MRI reconstruction",

abstract = "We introduce a 4-dimensional joint generative probabilistic model for estimation of activity in a PET/MRI imaging system. The model is based on a mixture of Gaussians, relating time dependent activity and MRI image intensity to a hidden static variable, allowing one to estimate jointly activity, the parameters that capture the interdependence of the two images and motion parameters. An iterative algorithm for optimisation of the model is described. Noisy simulation data, modeling 3-D patient head movements, is obtained with realistic PET and MRI simulators and with a brain phantom from the BrainWeb database. Joint estimation of activity and motion parameters within the same framework allows us to use information from the MRI images to improve the activity estimate in terms of noise and recovery.",

keywords = "Bayesian Networks, Emission Tomography, Molecular Imaging, Motion correction, Multi-modality",

author = "Stefano Pedemonte and Alexandre Bousse and Hutton, {Brian F.} and Simon Arridge and Sebastien Ourselin",

year = "2011",

doi = "10.1007/978-3-642-23623-5_73",

language = "English",

isbn = "9783642236228",

series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

pages = "581--588",

booktitle = "Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings",

edition = "Part 1",

note = "14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011 ; Conference date: 18-09-2011 Through 22-09-2011",

}

Pedemonte, S, Bousse, A, Hutton, BF, Arridge, S & Ourselin, S 2011, 4-D generative model for PET/MRI reconstruction. in Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings. Part 1 edn, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 6891, pp. 581-588, 14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011, Toronto, ON, Canada, 18/09/11. https://doi.org/10.1007/978-3-642-23623-5_73

4-D generative model for PET/MRI reconstruction. / Pedemonte, Stefano; Bousse, Alexandre; Hutton, Brian F. et al.
Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings. Part 1. ed. 2011. p. 581-588 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 6891).

Research output: Chapter in Book/Report/Conference proceeding › Conference proceeding contribution › peer-review

TY - GEN

T1 - 4-D generative model for PET/MRI reconstruction

AU - Pedemonte, Stefano

AU - Bousse, Alexandre

AU - Hutton, Brian F.

AU - Arridge, Simon

AU - Ourselin, Sebastien

PY - 2011

Y1 - 2011

N2 - We introduce a 4-dimensional joint generative probabilistic model for estimation of activity in a PET/MRI imaging system. The model is based on a mixture of Gaussians, relating time dependent activity and MRI image intensity to a hidden static variable, allowing one to estimate jointly activity, the parameters that capture the interdependence of the two images and motion parameters. An iterative algorithm for optimisation of the model is described. Noisy simulation data, modeling 3-D patient head movements, is obtained with realistic PET and MRI simulators and with a brain phantom from the BrainWeb database. Joint estimation of activity and motion parameters within the same framework allows us to use information from the MRI images to improve the activity estimate in terms of noise and recovery.

AB - We introduce a 4-dimensional joint generative probabilistic model for estimation of activity in a PET/MRI imaging system. The model is based on a mixture of Gaussians, relating time dependent activity and MRI image intensity to a hidden static variable, allowing one to estimate jointly activity, the parameters that capture the interdependence of the two images and motion parameters. An iterative algorithm for optimisation of the model is described. Noisy simulation data, modeling 3-D patient head movements, is obtained with realistic PET and MRI simulators and with a brain phantom from the BrainWeb database. Joint estimation of activity and motion parameters within the same framework allows us to use information from the MRI images to improve the activity estimate in terms of noise and recovery.

KW - Bayesian Networks

KW - Emission Tomography

KW - Molecular Imaging

KW - Motion correction

KW - Multi-modality

UR - http://www.scopus.com/inward/record.url?scp=80053529156&partnerID=8YFLogxK

U2 - 10.1007/978-3-642-23623-5_73

DO - 10.1007/978-3-642-23623-5_73

M3 - Conference proceeding contribution

C2 - 22003665

AN - SCOPUS:82255183364

SN - 9783642236228

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 581

EP - 588

BT - Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings

T2 - 14th International Conference on Medical Image Computing and Computer Assisted Intervention, MICCAI 2011

Y2 - 18 September 2011 through 22 September 2011

ER -

Pedemonte S, Bousse A, Hutton BF, Arridge S, Ourselin S. 4-D generative model for PET/MRI reconstruction. In Medical Image Computing and Computer-Assisted Intervention, MICCAI 2011 - 14th International Conference, Proceedings. Part 1 ed. 2011. p. 581-588. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). doi: 10.1007/978-3-642-23623-5_73

4-D generative model for PET/MRI reconstruction

Abstract

Publication series

Other

Keywords

Access to Document

Other files and links

Fingerprint

Cite this